| |
C-Reactive Protein, Statins, and Cardiovascular Risk: What Can JUPITER Teach Us?
|
| |
| |
Am Jnl of Kidney Disease
(May 2009)
Adam M. Rogers, MD, Michael G. Shlipak MD PhD
San Francisco VA Medical Center, University of California, San Francisco, CA
Commentary on Ridker PM, Danielson E, Fonseca FA, et al: Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med 359:2195-2207, 2008.
HMG-CoA reductase inhibitors (statins) have become a cornerstone therapy for cardiovascular prevention. Among participants both with and without established cardiovascular disease, placebo-controlled randomized clinical trials have demonstrated that statins reduce cardiovascular risk. These trials have steadily moved toward enrolling individuals at lower cardiovascular risk and with lower low-density lipoprotein (LDL) levels. The success of these trials has led to a vast adult population who are potential candidates for statins, possibly even including persons with normal LDL levels. Clinical dilemma and controversy have surrounded the selection of appropriate patients for statin therapy among the huge population at mild-to-moderate cardiovascular risk. JUPITER (Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin; ClinicalTrials.gov study number NCT00239681), published recently in the New England Journal of Medicine, is the latest randomized, double-blind, placebo-controlled trial to evaluate the role of statins for primary prevention of adverse cardiovascular outcomes.1
One potential test to identify the higher risk group among potential statin candidates is C-reactive protein (CRP), an inflammatory biomarker that is independently associated with cardiovascular risk.2, 3 Among persons at intermediate cardiovascular risk based on the Framingham risk score, higher CRP levels appear to identify some individuals at increased cardiovascular risk who might have a greater absolute benefit from statin therapy.4 Furthermore, statins lower CRP levels, which has been postulated to be an additional mechanism for their beneficial effects on cardiovascular risk beyond their effects on lipoproteins.5 Two important questions regarding the value of CRP as a clinical test that have not been addressed are: (1) would the use of CRP as a screening test to select statin candidates improve cardiovascular prevention; and (2) would monitoring CRP levels during statin therapy allow for more effective titration of statin dosing and thus improve outcomes. These key questions about CRP levels remain unanswered. The intriguing feature of JUPITER is that it bridges aspects of both a pharmaceutical trial and a study of a new diagnostic test. Below, we will attempt to disentangle these 2 issues relevant to JUPITER.
What Does This Important Study Show?
JUPITER differs from its predecessors in 2 major respects. First, the trial enrolled 17,802 apparently healthy individuals with low LDL cholesterol (< 130 mg/dL) in contrast with prior studies that recruited participants with elevated LDL cholesterol (> 130 mg/dL). Second, CRP was used as a screening criterion for entry into the trial among otherwise eligible candidates, and only patients with a CRP level greater than 2.0 mg/L were enrolled. The primary end point was a composite of myocardial infarction, stroke, arterial revascularization, hospitalization for unstable angina, or death from cardiovascular causes. The trial was stopped after a median follow up of 1.9 years due to a statistically significant decrease in the rate of the primary end point in the rosuvastatin group with a hazard ratio (HR) of 0.56 (95% confidence interval [CI], 0.46 to 0.69; P < 0.001). The benefits of rosuvastatin therapy persisted in all subgroups, including those with estimated glomerular filtration rate (GFR) less than 60 mL/min/1.73 m2, and remained highly significant for the combined end point of myocardial infarction, stroke, or confirmed death from cardiovascular causes (HR, 0.53; 95% CI, 0.40 to 0.69; P < 0.001). The number needed to treat to prevent 1 primary end point was calculated to be 95 over 2 years and 31 over 4 years. The rates of medication side effects were similar in the 2 groups, aside from an increased incidence of newly diagnosed diabetes in the rosuvastatin group (3.0% v 2.4%; P = 0.01). Despite these unequivocally positive findings, the authors do not offer specific recommendations regarding how broadly statin therapy should be expanded, or whether routine CRP testing should be adopted in order to optimize primary prevention.
How Does This Study Compare With Prior Studies?
JUPITER adds to an already extensive literature which largely supports the use of statins for both primary and secondary prevention of adverse cardiovascular and cerebrovascular events. After several early studies demonstrated the efficacy of statins in lowering LDL cholesterol, the Scandinavian Simvastatin Survival Study (4S), published in 1994, was a landmark trial of 4,444 patients with established coronary artery disease and elevated LDL cholesterol; the trial demonstrated an improvement in event-free survival among patients treated with simvastatin compared to placebo.6 Soon after, the 4S findings were confirmed by the Cholesterol and Recurrent Events (CARE) and Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) trials7, 8; on the basis of these studies, statins have been widely adopted as standard of care for ambulatory patients with coronary artery disease.
In parallel with secondary prevention studies, prior trials have also established the role of statins for the primary prevention of adverse cardiovascular outcomes. Published in 1995, the West of Scotland Coronary Prevention Study (WOSCOPS) randomly assigned 6,595 men with hypercholesterolemia (mean LDL level 192 mg/dL) but no history of coronary artery disease to pravastatin or placebo, and found significantly fewer adverse coronary events in the pravastatin group (relative risk [RR], 0.66; 95% CI, 0.59 to 0.75; P < 0.001).9 The Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TXCPS) in 1998 included men and women at lower cardiovascular risk and with lower LDL levels (mean LDL level 150 mg/dL), but found a similar relative risk (RR, 0.63; 95% CI, 0.50 to 0.79; P < 0.001).10 These primary prevention studies have found a similar relative benefit from statin therapy, but lower absolute risk reductions which is reflected in higher Ònumbers needed to treat.Ó JUPITER participants were at even lower risk than those in the prior primary prevention trials, so the overall health benefits of statins for cardiovascular risk reduction have now expanded to an even wider target population. However, at a public health level, the decision of whether or not to employ statins in a widespread prevention strategy for this target population must also factor in considerations of cost and potential long-term adverse effects from these medications.
What Should Clinicians and Researchers Do?
An interesting aspect of JUPITER for nephrologists might be how the trial can be related to other novel biomarkers under investigation or in clinical use in nephrology and other medical fields. High-sensitivity CRP level greater than 2.0 mg/L was used as an inclusion criterion in JUPITER, and CRP level was shown to be 37% lower in the statin group at 12 months compared to the placebo group.1 However, this trial did not evaluate whether or not screening with CRP levels was beneficial, nor did it evaluate the value of monitoring CRP levels among patients treated with statins. Although CRP levels are predominately featured in the study's title, design, and presentation, JUPITER was a statin trial, not a CRP trial. Because many new biomarkers in kidney disease are in development, JUPITER offers us an opportunity to contrast trials of diagnostic tests versus trials of therapeutic interventions.
In the opening section, we alluded to 2 potential trial designs to evaluate a new diagnostic test and we noted that neither has been used to evaluate CRP. These designs are: (1) the test as a screening strategy to select candidates for treatment, and (2) the test as a tool to guide therapeutic intensity. For the first design, we will use the example of a new urine biomarker of acute kidney injury (AKI), neutrophil gelatinase-associated lipocalin (NGAL). The goal of new AKI biomarkers is the early detection of AKI, so that an existing or new AKI therapy can be initiated at the earliest possible time. A tempting design for a future clinical trial would be to screen patients at high risk of AKI for elevated NGAL levels; those with high NGAL would be randomly assigned to a novel therapy versus usual care. This design, however, would mirror JUPITER; it would not prove the value of NGAL for AKI detection any more than JUPITER demonstrated the value of CRP. To evaluate NGAL as a screening test, we would need to assign 2 large groups of patients to the following strategies: (1) usual detection of AKI and usual AKI therapy, versus (2) screening all patients with NGAL and treating positive patients at the earlier time-point allowed by NGAL. This strategy isolates the value of the diagnostic test and allows us to determine the actual costs of employing the screening strategy in the target population.
The second trial design involves evaluation of the biomarker as a monitoring strategy for determining treatment effectiveness over time; 1 group of patients has frequent test measures which are reported to the treating clinicians and the other group receives usual care. A useful example is the recently published Trial of Intensified Versus Standard Medical Therapy in Elderly Patients With Congestive Heart Failure (TIME-CHF), which randomly assigned patients with heart failure and elevated N-terminal brain natriuretic peptide (BNP) levels to either symptom-guided medical therapy or intensified BNP-guided therapy added to symptom control.11 This study design effectively isolates the role of BNP monitoring by focusing on the additive effect of BNP monitoring compared with standard of care alone, and allows us to interpret the clinical relevance of following BNP levels as part of in-hospital heart failure management. In TIME-CHF, BNP-guided therapy was not significantly different than symptom-guided therapy for the primary outcome of the study. An alternative JUPITER trial design could have followed this model to address the utility of following CRP levels to guide the intensity of statin treatment.
The publication of JUPITER was followed by extensive discussion in both the lay press and scientific literature about whether or not CRP levels should become a routine clinical measure. Although CRP is not an expensive test, it is important to distinguish a trial that is predominantly a pharmaceutical study (JUPITER) with a trial that is actually designed to evaluate a diagnostic test (TIME-CHF). As exciting new nephrology biomarkers move closer to clinical application-injury biomarkers, new filtration biomarkers, or imaging techniques-their efficacy for improving health outcomes should be evaluated and tested rigorously. Use as a criterion for entry into a clinical trial is not sufficient evidence to justify widespread deployment of a new diagnostic test.
Acknowledgements
Financial Disclosure: None.
|
|
| |
| |
|
|
|
